Inspection and Testing of Failed Tow Wire Rope PDF

TTI Testing Ltd
Unit 2, Hithercroft Road, Wallingford, Oxfordshire, OX10 9DG, UK
www.tti-testing.com
Inspection and testing of failed tow wire rope

from tug ALP Forward
FINAL REPORT
TTI Reference number TTI-IMLR-2016-5284
Date Rev. Description Prepared by Authorised by

06/01/17 0 for issue to client
19/03/17 1 Following review
Distribution: Attention:
TTI Testing Ltd (author, file)
MAIB
ALP Forward wire rope failure - Final Report
Summary
This report presents the findings of a study undertaken on a wire rope

being used as the towing wire on ALP Forward to tow the Transocean
Winner between Norway and Malta when it parted in heavy weather on 8th
August 2016. The platform Transocean Winner subsequently ran aground
at the Isle of Lewis.
The MAIB wished to understand why the wire rope failed and why it failed
where it did.
Initial work was undertaken which inspected sections of the wire rope from
the following locations along its length:
• a 5 m sample from the drum ('drum sample') which had not previously
been deployed and was considered to be representative of the rope's
new condition;
• Approx. 20 m samples from each side ('tug side' and 'rig side') of the main
wire rope failure;
• a 20 m sample from adjacent to the socket which connected the wire
rope to the towing bridle ('bridle sample').
Initial visual inspection of the tug side sample and bridle sample found
numerous shear type wire breaks in the outer strand outer wires. These
were located in the valley positions between outer strands and thought to
be caused by heavy contact between the strands.
Full dismantling inspection and testing was made of selected 1 m long
samples from each of the drum, tug and bridle sections. Testing of wires
from the drum sample indicated that the rope as purchased was
manufactured from wire stock with mechanical properties which satisfied
the requirements of API 9A.
The tug and bridle samples were in similar condition, and considered to be
representative of the condition of the length of wire rope deployed and so
of that at the main failure at the time of the incident:
• The lubricant level was low;
• The outer strands were covered in a light coating of corrosion;
• There was very little protective galvanising remaining on the exposed
surface of the wires;
• Several outer strand outer wire breaks were noted along the length; and,
• The independent wire rope cores were in very poor condition, each with
over 100 wire breaks on the 1 m samples.
It is impossible to know when all this degradation occurred, but difficult to
see it all being as a result of the job which ALP Forward was engaged at the
time of the incident.
i
Tensile breaking load tests were undertaken on whole rope samples from
the drum, tug and bridle sections. The results of these tests showed that
the strength of the drum sample was down by about 7% on the originally
measured breaking strength, whilst the tug and bridle samples were both
over 20% down on as new.
The nature and distribution of the wire failures in the breaking load
samples was very similar to the main failure (tug side sample). The outer
strand outer wires had predominantly failed in shear, whilst the outer
strand inner wires had failed in classic cup and cone type fractures. (The rig
side sample main failure was not inspected as it had suffered extensive
abrasion damage dragging along the seabed post failure.)
A consideration of the fatigue performance of the six strand rope indicated
that during the storm the wire rope would have accumulated fatigue
damage which would contribute to degradation.
Analysis of catenary behaviour indicated that under high tensions the wire
rope would be reliant on elastic stretch to accommodate any sudden
displacement caused by waves, and that very high tensions could easily be
produced by relatively small displacements.
It is concluded that the wire rope was weakened by accumulated damage
during the storm and ultimately failed by tensile overload in its weakened
state.
ii
Contents
Glossary .............................................................................................................................. v
1 Introduction ................................................................................................................ 1
2 Main tow wire rope..................................................................................................... 2
3 The towing arrangement and sample location ............................................................ 4
4 Delivery of samples ..................................................................................................... 7
4.1 Delivery of the drum sample ....................................................................................... 7
4.2 Delivery of the tug side and rig side samples .............................................................. 8
5 Initial inspection of samples ...................................................................................... 11
5.1 Initial inspection of drum sample .............................................................................. 11
5.2 Initial inspection of tug side sample .......................................................................... 12
5.3 Initial inspection of rig side sample ........................................................................... 15
6 Initial cutting schedule .............................................................................................. 18
6.1 Cutting schedule for drum sample ............................................................................ 18
6.2 Cutting schedule for tug side sample ........................................................................ 18
6.3 Cutting schedule for rig side sample ......................................................................... 18
7 Delivery of bridle sample........................................................................................... 20
8 Initial inspection of additional sample....................................................................... 21
8.1 Initial inspection of bridle sample ............................................................................. 21
8.2 Proposed cutting schedule for bridle sample............................................................ 22
9 Wire rope breaking strength measurement............................................................... 23
9.1 Test equipment.......................................................................................................... 23
9.2 Results of the breaking load tests ............................................................................. 25
10 Strip down inspection of 1 m samples ....................................................................... 31
10.1 Strip down inspection of Drum Sample ................................................................. 31
10.2 Strip down inspection of the tug side sample ....................................................... 40
10.3 Strip down inspection of the bridle sample........................................................... 48
10.4 Measurements made on strip down samples ....................................................... 57
iii
11 Inspection of tug side failure ..................................................................................... 58

12 Mechanical tests on individual wires ......................................................................... 61
12.1 Tensile tests ........................................................................................................... 63
12.2 Torsion tests ........................................................................................................... 64
12.3 Reverse bend test .................................................................................................. 64
13 Wire zinc coat weight ................................................................................................ 65
14 Results of the mechanical tests ................................................................................. 66
15 Results of the zinc coat weight measurement ........................................................... 71
16 Catenary behaviour ................................................................................................... 75
16.1 General discussion of catenary behaviour............................................................. 75
16.2 Approximate calculation of maximum loads in the towing wire rope .................. 78
16.2.1 Characterisation of the wire rope ...................................................................... 78
16.2.2 Definition of the model ...................................................................................... 79
16.2.3 Calculation of peak loads ................................................................................... 80
16.3 Fatigue loading ....................................................................................................... 81
17 Discussion and Conclusions ....................................................................................... 83
18 References ................................................................................................................ 86
Appendices ....................................................................................................................... 87
Appendix A - Tow wire inspection certificate .................................................................... 88
Appendix B - Calibration certificate for the 14,000 kN test machine ................................. 90
Appendix C - Certificates for the break load tests ............................................................. 96
Appendix D – Calibration certificate for 250 kN tensile machine ....................................... 99
Appendix E - Results of the mechanical testing on rope wires ..........................................102
Appendix F - Results of the average zinc coat weight measurement ................................112
iv
Glossary
ABL (measured) Actual Breaking Load

d Rope nominal diameter
dw Wire diameter
Fmin Minimum breaking load
IWRC Independent Wire Rope Core
Lay length the axial length over which a wire (or strand) make one helical revolution
about a strand (or rope).
RHO Right Hand Ordinary lay
v
1 Introduction
TTI Testing (TTI) has been contracted by the Marine Accident Investigation Branch (MAIB)
for assistance in determining the cause of a wire rope failure (MAIB Reference number
4/3/144). The wire rope was being used as the towing wire on ALP Forward to tow the
Transocean Winner between Norway and Malta when it parted in heavy weather on 8th
August 2016. The platform Transocean Winner subsequently ran aground at the Isle of
Lewis.
The MAIB seeks to understand why the wire rope failed and why it failed where it did.
Specifically:
• The condition of the wire rope where it failed and at other selected points
• The implications of its condition on serviceability at the time of failure
• The mechanism of the wire rope’s failure
• The load on the wire rope when it failed
• The load the wire rope would have failed at had it been new
As a secondary issue, the emergency towing arrangement on Transocean Winner included a

steel pennant, connected at one end to the rig and at the other to a floating polypropylene
(PP) pick-up rope. The PP rope was in turn connected to a messenger line, and the outboard
end of the messenger was connected to a wave-rider (Norwegian buoy). When Transocean
Winner was examined after grounding, a section of the messenger line and buoy were no
longer present. The MAIB wished to understand how the messenger line came detached
from the buoy. In the absence of other evidence, this could only be determined by a
thorough examination of the end of the messenger line.
This document presents background information relevant to the study, and describes the
initial inspection and testing of the sections of wire rope which were sent to TTI Testing,
Wallingford. Results of the tests are presented along with a discussion of the results to
address the questions raised above.
The examination of the pick-up rope and messenger line is covered in a separate report.
1
Table 2.2 presents data from the individual wire tests for the wires in the main load bearing
strands of the rope.
Wire diameter, Tensile strength Bend (Average) Torsion Weight of zinc

dw (Average) (Average) coating
(Average)
[mm] [N/mm2] [-] [turns /100 dw] [g/m2]
4.39 2164 N/A 13 161
2.73 2096 N/A 23 142
3.53 2128 N/A 17 146
3.62 2075 N/A 16 146
4.81 2087 N/A 12 163
Table 2.2: Average wire properties for the main load bearing outer strand wires (from Usha Martin
Inspection Certificate - See Appendix A).
3
Figure 3.2: Schematic diagram of the bridle arrangement connecting Transocean Winner to ALP
Forward.
Figure 3.3: View on failure of 77 mm wire rope (tug side) (Photograph supplied by MAIB).
5
Figure 3.4: Tug side sample taken for inspection: 20 m from the failure towards the tug ALP Forward
(Photograph supplied by MAIB).
6
4 Delivery of samples
4.1 Delivery of the drum sample

The drum sample was delivered to TTI Testing from Rotterdam in a sturdy wooden crate
which had been made especially for the sample (Figures 4.1 and 4.2).
Figure 4.1: Drum sample as received at TTI Testing.
Figure 4.2: Drum sample as received at TTI Testing with crate lid fully removed.
7
4.2 Delivery of the tug side and rig side samples

The tug side and rig side samples were transported to TTI Testing in a locked 10' container
(Figure 4.3). The two samples were coiled and separated from each other by sheets of
10 mm plywood and tarpaulins (Figures 4.4 and 4.5).
The samples were carefully removed from the container in turn and laid out on the
laboratory floor to permit an initial inspection (Figure 4.6).
Figure 4.3: Container with tug side and rig side samples as received at TTI Testing.
8
Figure 4.4: Container opened to show tug side (top) and rig side (bottom, wrapped) as received.
Figure 4.5: Rig side sample in container (unwrapped).
9
Drum sample
Rig side sample
Tug side sample
Figure 4.6: Samples laid out in TTI Testing laboratory for initial inspection.
10
5 Initial inspection of samples
5.1 Initial inspection of drum sample

Figure 5.1 shows the general condition of the drum sample as received. Generally the
sample was well lubricated particularly in the outer strand valleys. Where present the
lubricant was glossy and black in appearance, and as new. There were also some grey
deposits of what were thought to be mud or silt. It is not known at what stage these were
deposited on the wire rope. Superficial corrosion was noted on the surface of the wires.
Mid-length along the sample was found an area of crushing damage on one of the strands.
Whilst not ideal, it is thought that this damage would not adversely affect the breaking
strength of the sample as the damage should 'pull out' under the high loads.
Figure 5.1: General condition of the drum sample as received.
Figure 5.2: Area of crushing damage found on the drum sample (mid-length).
11
It is immediately noted that the lubricant between the strands is missing, the protective
galvanising seems low or lost, and the wires (the outside of the wire rope) are covered in
what appears to be superficial corrosion. Wire breaks were noted distributed along the 10 m
length, both along and around the wire rope circumference. In all 40 visible breaks were
counted (there may well have been more), all of which initiated at the inter-strand contact
region of the wire rope and appeared to be shear type in nature. Beyond 10 m from the cut
end there were many more wire breaks.
A more detailed assessment of the condition of the tug side sample was made on the full
inspection of a 1 m sample, including mechanical tests on the wires.
Figure 5.7: Outer strand wire break (valley) measured at 1.4 m from cut end on tug side sample.
Figure 5.8: Two of six outer strand wire breaks measured at 2.0 m from cut end on tug side sample.
14
Figure 5.9: Two outer strand wire breaks measured at 5.6 m from cut end on tug side sample.
5.3 Initial inspection of rig side sample

The rig side sample was uncoiled and measured at 23.2 m overall length, although 9 m of
that length comprised the remains of the single strand which was noted to have failed at
10.2 m on the tug side sample.
Comparison of Figures 4.4 and 4.5 shows that the tug side and rig side samples appeared
quite different. (It is noted that the samples had been treated or coated post recovery to
preserve the evidence which may contribute in part to this disparity.) The tug side sample
had a uniform coating of corrosion, whilst the rig side sample had marked shiny patches on
the crown positions of the strands, the most heavily affected area being in the first 4 m or so
from the cut end. Figure 5.10 shows a general view of the sample near the cut end. It may be
seen that there are valley wire breaks as were noted on the tug side sample, but also
significant abrasion damage (Figure 5.11).
The lack of corrosion on the surface of the abrasion suggests that it is very recent. It must be
considered possible that this damage (the abrasion) was caused to the wire rope after the
failure of the tow. Given the position of the main failure along the wire rope the failed end
of the wire rope would have dragged along the seabed for some distance (≈ 3 miles) as the
rig drifted. The nature of the seabed is not known, however small pieces of stone were
found between strands in positions along the length of the sample (see for example, Figure
5.12).
The suggestion that the abrasion damage occurred post to failure is supported by the
condition of the IWRC which had displaced from the centre of the wire rope construction at
5.7 m from the cut end (Figure 5.13).
Figure 5.14 shows the failed end of the sample. The end of the IWRC was measured as
12.1 m from the cut end. It is therefore quite possible that a short section is missing. This is
not surprising given the condition of the core noted in other positions along the wire rope
which showed numerous wire breaks.
15
Figure 5.10: General condition of the wire rope about 1.1 m from cut end of rig side sample.
Figure 5.11: Significant abrasion on the surface of the outer wires 2.0 m from cut end of rig side
sample.
Figure 5.12: Small pieces of stone noted trapped in the wire rope at 5.6 m from cut end of rig sample.
16
6 Initial cutting schedule

An initial cutting schedule was drawn up based on the on the preliminary inspection of the
three samples which were supplied (Figure 6.1).
6.1 Cutting schedule for drum sample

• A 1 m section was taken from the end of the drum sample for a full strip down
inspection and mechanical tests on representative wires (the strip down inspection).
• The remaining 4 m length was sent to Mennens for whole wire rope breaking
strength measurement. It is noted that this section included the area with crushing
damage noted in §5.1, but it was not thought that this would affect the strength
measurement. (In any case, as the damage was mid length on the sample it was
impossible to avoid.)
6.2 Cutting schedule for tug side sample

• The sample for break load was taken as the first four meters from the cut end.
• A further 1 m sample for the strip down inspection and mechanical tests on
representative wires was taken at 8 m - 9 m from the cut end. This section was
chosen as it was close to the start of the main failure whilst avoiding major
disturbance to the construction.
6.3 Cutting schedule for rig side sample

The rig side sample was cut at 4 m and 5 m from the cut end. This provided samples for the
breaking load and strip down inspection. However, as discussed in §5.3 above, it seemed
unlikely that this sample was representative of the wire rope condition at time of the failure.
Discussion with the MAIB identified another section of the wire rope which was available - a
section of rig side wire rope from near to the bridle which would have experienced the same
loading, but not have dragged on the sea bed. This sample was termed the 'bridle sample'.
18
7 Delivery of bridle sample

The bridle sample was delivered to TTI Testing from Montrose coiled in a 10 ft half height
container, protected by a tarpaulin. It was carefully removed from the container and placed
in the laboratory for inspection (Figure 7.1).
Figure 7.1: Bridle sample as received at TTI Testing complete with closed spelter socket.
20
8 Initial inspection of additional sample
8.1 Initial inspection of bridle sample

Figure 8.1 shows the general condition of the bridle sample as received. In general terms the
sample appeared to be in the same condition as that of the tug side sample (compare with
Figures 5.7 - 5.9):
• the lubricant on and between strands was missing;
• the protective galvanising seems low or totally lost, and the surface of the wire rope
is at the point of onset of corrosion.
Wire breaks were observed along the sample (40 were counted over the 20 m length). Two
groups of wire breaks were observed at 15.2 m (eight), Figure 8.2, and at least six at 16.5 m,
Figure 8.3. (All distances measured from the front of the socket face.) In all eighteen wire
breaks were counted over a 30d1 length (2.31 m).
As with the tug side sample, all the observed wire breaks had initiated at the inter strand
contact position. Wire breaks in this position can be very difficult to identify, and it is quite
possible that there are many more in this section of the wire rope.
Figure 8.1: General condition of the bridle sample as received (7.1 m).
Figure 8.2: Group of at least eight wire breaks noted at 15.2 m on the bridle sample.
1
d is the nominal diameter of the rope and used as a characteristic length in the definition of discard criteria.
21
Figure 8.3: Group of at least six wire breaks noted at 16.5 m on the bridle sample (the blue paint is
from the shipping container).
8.2 Proposed cutting schedule for bridle sample

It was considered that the bridle sample was a fair representation of the condition of the
wire rope at the time of failure. Reference to the main tow-wire log [4] indicates that all the
recorded tow miles were made with at least 200 m of wire rope paid out.
The following cutting schedule for the bridle sample was employed, based on the
preliminary inspection (see also Figure 6.1).
• A 1 m length at 7 m - 8 m for a full strip down inspection and mechanical tests on

representative wires (the strip down inspection). This left a 7 m sample including the
socket should there be any wish to conduct a break strength measurement on the
fitting/wire rope assembly.
• A 4 m length at 13.5 m - 17.5 m which was sent to Mennens for whole wire rope
breaking strength measurement. It is noted that this section included the 18 wire
breaks over a length of 30d noted in §8.1 above.
22
9 Wire rope breaking strength measurement
9.1 Test equipment

The tensile testing equipment used to conduct the breaking load tests was at Mennen
Dongen BV in The Netherlands. This equipment (Figure 9.1) may be used for testing at loads
up to 14,000 kN (14 MN or 1,400 tonnes). Mennens is a ISO 9001 certified company, and the
test machine is calibrated to Class 1 according to ISO 7500. Appendix B presents a calibration
certificate for this equipment.
High speed camera filming of the breaking load test on the samples was undertaken to
examine whether there was any difference in the failure mode of the three samples. Figure
9.1 shows the camera being set up, and Figure 9.2 the camera and all the lighting ready for
the test.
Figure 9.1: Wire rope sample ready for testing in Mennens' 1,400 tonne tensile testing machine.
23
Figure 9.2: Wire rope sample ready for testing with high speed filming equipment in place.
24
9.2 Results of the breaking load tests

Table 9.1 presents the results of the breaking strength measurements conducted on the
three samples. Appendix C presents the Mennens tests reports for these tests.
Sample Breaking load [kN] Breaking load [%Fmin] Breaking load [%ABL]
Drum 4,493 94.4 92.7
Tug side 3,815 80.2 78.7
Bridle 3,707 77.9 76.5
Table 9.1: Results of the breaking strength measurement on the three wire rope samples.
Figures 9.3 - 9.5 show the drum sample following testing. It may be seen that the sample
failed clear of the terminations (Figures 9.3 and 9.4). Examination of the failure showed that
the majority of the wire failures were from tensile overload (as would be expected), with a
few shear type failures (given the helical nature of wire rope, this is also as expected).
Figure 9.3: Drum sample in test machine after testing.
25
Figure 9.4: Drum sample failure clear of termination.
Figure 9.5: Drum sample detailed view on failure region.
26
Figure 9.6 shows the broken wires which fell out of the tug side sample when it was
terminated for testing. The wires are largely from the IWRC, which as the initial inspection
had shown was in fairly poor condition. In addition to the IWRC wires there were also some
outer strand outer layer wire breaks. The tug side sample failed mid-length (Figure 9.7).
Figure 9.6: Tug side sample - broken wires found in preparation of sample terminations. (Mostly
IWRC wires, but also some outer strand outer wires).
Figure 9.7: Tug side sample after testing, failure mid-length.
27
Figures 9.8 and 9.9 show views on the groups of wire failures after testing of the tug side
sample. The inner layers of the main strands may b readily identified by their paler colour. It
can be seen that these wires failed with tensile cup and cone type failures. There were also
some shear failures (e.g. centre of Figure 9.9).
Figure 9.8: Tug side sample after testing, detailed view on wire failures.
Figure 9.9: Tug side sample after testing, detailed view on wire failures.
28
As with the tug side sample, the core of the IWRC was found to be mostly in pieces when the
sample was prepared for testing (Figure 9.10). Additionally, there were also breaks in some
outer strand outer wires. As with the other samples, the bridle sample failed clear of the
terminations (Figure 9.11).
Figure 9.12 shows a group of shear wire breaks on one of the outer strands after testing. This
view is interesting as it shows clearly that there were three existing wire breaks in the
sample (corroded fracture surfaces), and two wires which were broken in the breaking load
test (fresh ends).
Figure 9.10: Broken wires found in preparation of Bridle sample terminations - mostly IWRC wires.
Figure 9.11: Bridle sample after testing, failure mid-length.
29
Figure 9.12: Bridle sample after testing, detailed view on wire failures (old - rusty ends, new - shiny).
30
10 Strip down inspection of 1 m samples

Sections 1 m long were selected for visual inspection and mechanical testing of
representative wires from across the wire rope construction. As noted above, the rig side
sample had suffered too much post failure damage to permit a fair assessment of the
condition of the wire rope at the time of the incident, thus the following samples were
inspected:
• Drum sample;
• Tug side sample; and,
• Bridle sample.
The visual inspection of the samples was in line with the criteria set out in BS ISO 4309:2010
[5], and paid particular attention to corrosion, lubrication and mechanical damage.
Tests were performed on wires taken from positions throughout the wire rope cross section
to assess:
• Tensile strength;
• Torsions to failure;
• Reverse bends to failure; and,
• Residual zinc coat weight.
10.1 Strip down inspection of Drum Sample

Figure 10.1 shows the general condition of the drum sample. The crown position of the
sample was lacking in lubricant, and there were areas of white zinc oxide associated with
loss of galvanic protection. In some areas light corrosion was noted (Figure 10.2). There were
also patches of what appeared to be mud. It is not known when this was deposited on the
wire rope surface.
In the valley positions between strands the lubricant level was very good, but it was hard
and glassy in appearance. It is noted that the wire rope inspection certificate states that the
lubricant is 'D / bitumin lube', so it is likely that this lubricant is the original as manufactured.
Figure 10.3 shows the sample with two outer strands removed to reveal the IWRC. It may be
seen that the lubricant level on the IWRC is very low, and what lubricant there is, is dry.
31
Figure 10.1: General condition of the drum sample.
Figure 10.2: Detailed view on the drum sample showing an area of corrosion. Note also the 'mud
deposits' and the good level of lubricant in the valleys.
32
Figure 10.3: Drum sample with two outer strands removed to show the condition of the IWRC - the
lubricant level is very low. Note also the light corrosion on the outer strand (top right)
and the glassy bitumous lubricant in the valley positions (of the outer strands).
Figure 10.4 shows an outer strand after removal from the wire rope sample and partially
cleaned. Sections of lubricant have been left in place to highlight the difference in residual
galvanising levels between the section of strand normally on the outside of the wire rope,
and that facing towards the IWRC.
Figure 10.4: Drum sample outer strand partially cleaned to show the difference in residual
galvanising levels for sections on the inside and outside of the wire rope.
33
The outer strand was then stripped down layer by layer to permit assessment of the internal
condition. Figures 10.5 - 10.7 show the second and third layers and core wire of the outer
strand. It may be seen that the inner layers and core of the outer strand are well lubricated
and have retained a good galvanising covering. No signs of degradation or wear were noted.
before cleaning
after cleaning
Figure 10.5: Drum sample outer strand - second layer of wires.
34
before cleaning
after cleaning
Figure 10.6: Drum sample outer strand - third layer of wires.
before cleaning
35
after cleaning
Figure 10.7: Drum sample outer strand - third layer of wires and core wire (bottom).
Figure 10.8 shows the Manufacturers' marker tape which was found along the outside of the
IWRC. It confirms that the wire rope was manufactured by Usha Martin.
Figure 10.8: Manufacturers' identification tape found under the outer strands and on the IWRC
(Drum sample).
Figure 10.9 shows the drum sample with all the outer strands removed to show the IWRC. As
noted above, the lubricant level is low and what is there appears denatured and powdery.
An outer strand was removed to permit further examination. Figure 10.10 shows the core
strand after cleaning. It may be seen that the residual galvanising is patchy, but there is no
corrosion as such on the core strand. Dismantling the core strand shows that the lubricant
inside the strands is good and sticky, and that the residual zinc is a little patchy, but overall
good (Figure 10.11).
36
Figure 10.9: Drum sample IWRC.
Figure 10.10: Drum sample IWRC outer strand cleaned.
37
before cleaning
after cleaning
Figure 10.11: Drum sample IWRC outer strand outer wires and core (bottom).
Figure 10.12 shows the IWRC core strand, which is similar in appearance to the IWRC, but
with more lubricant. On cleaning the outer surface of the core (Figure 10.13) it may be seen
that the residual galvanising is also patchy, but as with the outer strands, there is no
corrosion. Figure 10.14 shows the core after dismantling. It may be seen that the lubricant
level is good, and appears sticky in nature. The galvanising is a little patchy on the more
exposed surfaces.
Figure 10.12: Drum sample IWRC core strand.
Figure 10.13: Drum sample IWRC core strand cleaned.
38
before cleaning
after cleaning
Figure 10.14: Drum sample IWRC outer wires and core strand (bottom) cleaned.
Following inspection of the drum sample all six outer strands and the IWRC were stripped
down to check for wire breaks. None were found.
Wires were selected from various positions in the wire rope construction for the
measurement of the level of residual galvanising as well as tension, torsion and reverse
bending tests. The results of these tests will be presented later in this report.
39
10.2 Strip down inspection of the tug side sample

Figure 10.15 shows the general condition of the tug side sample in the 'as received'
condition. Its appearance is quite different to that of the drum sample (cf. Figure 10.1). The
lubricant level was low to none, and there was a covering of light corrosion along the length.
A single wire break was noted in one of the outer strands in the valley position, that is at the
position where the outer strands contact one another. Figure 10.16 shows a detailed view
on the wire break which had a 45° shear failures fracture surface typical of combined axial
and transverse loading (axial load in the wire rope combined with transverse load as two
strands press together). Note also that there is very little if any residual galvanising and the
corrosion which has just started to roughen the surface of the wires.
Figure 10.15: General condition of the tug side sample before cleaning.
Figure 10.16: Detailed view on single wire break noted in valley position on tug side sample.
40
Figure 10.17 shows the tug side sample with two outer strands removed to reveal the IWRC.
The lubricant level was generally low (a single large patch was noted in one area (Figure
10.18)). As with the outer strands the residual galvanising was low, and there was corrosion
along the length of the core. Many wire breaks were noted in the IWRC outer strands
(Figures 10.17 - 10.20). The remaining outer strands were carefully removed to permit full
inspection of the core (Figure 10.18).
Figure 10.17: Tug side sample with two outer strands removed to reveal IWRC.
Figure 10.18: Tug side sample, general condition of IWRC. Five wires dropped out of the construction
as the outer strands were removed. Some areas of good level, good nature lubricant,
but on the whole the lubricant level was low and dry.
41
Figure 10.21 shows the IWRC with two strands removed to show the core. The lubricant
level was much better than on the outer strands, but it was dry and flaky (Figure 10.22).
There did not appear to be much corrosion - the main feature noted was that the whole core
strand was broken through in several places. Once the outer strands were all removed it
could be seen that the core strand was broken through completely in five places (Figure
10.23), as well as having other wire breaks along its length.
Figure 10.21: Tug side sample, IWRC with two strands removed to show core. Core strand broken
through. Good level of lubricant.
Figure 10.22: Tug side sample, dry lubricant which dropped out of the wire rope.
43
Figure 10.23: Tug side sample IWRC core - broken through in five places (6 pieces) over 1 m length.
Figures 10.24 and 10.25 show sections of the IWRC core and an outer strand after cleaning.
The IWRC core has some residual galvanising present, whilst the core strand has very little if
any. The corroded appearance noted above (Figure 10.21) has cleaned off. The wire failures
in both the strand and the core are the shear type, which are to be expected as the IWRC
experiences high transverse loads as it supports the main outer strands in the wire rope.
Figure 10.24: Tug side sample IWRC core strand after cleaning.
Figure 10.25: Tug side sample IWRC outer strand after cleaning.
44
Figure 10.26 shows one of the main outer strands after cleaning. The wire break was located
in the valley position in the wire rope and was not detected until the strands were removed
to inspect the core. (In all four outer strand outer wire breaks were found in the 1 m sample
- only one of which was found before the wire rope was dismantled.) The break has initiated
at the wear patch on the surface of the wire, and failed in a mixture of shear and fatigue.
The 'beach' marks characteristic of fatigue may be seen on the right hand edge of the
fracture surface. It is also noted that there is very little residual galvanising left on the
surface of the strand.
Figure 10.26: Tug side sample main outer strand after cleaning.
Figure 10.27 shows an outer strand with four outer wires removed to reveal the second
layer. The contrast between the outer layer and second layer is quite marked: the level of
lubricant in the second layer is better, and although still low, the residual galvanising seems
better. Figure 10.28 shows the third layer of wires in the main strand before and after
cleaning. This 'far' into the strand the lubricant is much better both in level and nature and
the residual galvanising is also higher. Similarly the strand core wire had very good lubricant
and high residual galvanising levels. No corrosion was noted in any of the inner layers of the
outer strand inspected.
Following inspection of the sample, all outer strands were completely dismantled to check
for wire breaks. Apart from the four outer strand outer wire breaks, no internal wire breaks
were found. Indeed, the inside of the main strands were in generally good condition.
In contrast, as noted above, the IWRC was in very poor condition. The strands and core were
stripped down to count the wire breaks. In all there were 62 breaks in the outer strands and
59 in the IWRC core (total 121 wire breaks). Obviously there will be a loss of strength
associated with this level of degradation: the IWRC of a six strand wire rope typically carries
about 15% of the wire rope strength. What is more important is that a core with this level of
damage will not perform its main function which is to support the outer strands so that as it
bends they are free to slide and do not contact one another.
45
before cleaning
after cleaning
Figure 10.27: Tug side sample, main outer strand with four outer layer wires removed to expose the
second layer.
46
before cleaning
after cleaning
Figure 10.28: Tug side sample, main outer strand third layer of wires.
before cleaning
after cleaning
Figure 10.29: Tug side sample, main outer strand core wire.
47
10.3 Strip down inspection of the bridle sample

Figure 10.30 shows a general view of the bridle sample. It is fairly similar in appearance to
the tug side sample: the external lubricant has gone and the surface of the wire rope is
covered with a light corrosion which has just started to cause a roughening of the wire
surface (Figure 10.31). Seven outer strand outer wire breaks were counted along the length
(four of which may be seen in Figure 10.30). All wire breaks were in the valley positions and
shear type in appearance.
Figure 10.30: General view on bridle sample.
Figure 10.31: Detailed view on bridle sample - note no visible galvanising and a uniform covering of
corrosion causing roughening of the surface of the wires.
48
Figures 10.32 and 10.33 show the bridle sample with two outer strands removed to expose
the IWRC. It may be seen that in some areas the lubricant was a quite a high level, if
somewhat denatured (Figure 10.32), whilst in other places along the core the level was low
(Figure 10.33).
Figure 10.32: Bridle sample with two outer strands removed - area of good lubrication on IWRC.
Figure 10.33: Bridle sample with two outer strands removed - area of low lubrication on IWRC.
49
The outer strands were all removed to permit a detailed inspection of the core
(Figure 10.34). In general the lubricant level was higher than the tug side sample, making it
harder to assess for wire breaks (the sample was subsequently dismantled to allow a full
count to be made). It was also noted that there was sand on the surface of the IWRC. It is
not clear how this came to be on the wire rope, particularly in this position in the wire rope
construction, but with the lubricant missing from between the outer layer of strands it
would be possible for the sane to enter and become trapped in the wire rope.
Figure 10.34: Bridle sample IWRC. Note the lubricant level is reasonable if a little dry and
denatured, also the grains of sand.
Figure 10.35 shows the IWRC with three outer strands removed. The condition is very similar
to the tug side sample. The lubricant is at a reasonable level, if a little dry. As with the tug
side sample the core strand was broken through in several (in this case four) places
(Figure 10.36).
The wire failures in the core strand were typically shear type with the characteristic 45°
fracture surface (Figure 10.37). Figure 10.38 shows a dismantled section of core strand
before and after cleaning. It can be seen that the residual galvanising level on the wires is
very low, with only a few small patches observed, however the wires do not appear to be
corroded.
Figure 10.39 shows a IWRC strand after removal from the core and cleaning. The wire breaks
are typical of those found in the core strands. As with the tug side sample, all strands and
the core were stripped down to count the number of breaks. In the IWRC outer strands 69
wire breaks were counted and in the core 48 (total 117 breaks in 1 m of IWRC).
Figure 10.40 shows a section of dismantled IWRC strand before and after cleaning. As with
the core strand the residual galvanising is very low. Additionally, some areas of the wire
surface show a roughening from the start of corrosion.
50
Figure 10.35: Bridle sample IWRC with three outer strands removed to show the core (all core
strand wires broken).
Figure 10.36: Bridle sample IWRC with three outer strands removed to show the core (all core
strand wires broken).
Figure 10.37: Bridle sample IWRC core showing typical shear type wire failures.
51
before cleaning
after cleaning
Figure 10.38: Bridle sample IWRC core, (some) outer wires and core wire (bottom).
Figure 10.39: Bridle sample IWRC outer strand (after cleaning).
52
before cleaning
after cleaning
Figure 10.40: Bridle sample IWRC strand, (some) outer wires and core wire (bottom).
53
Figure 10.43 shows the second layer of the main strand before and after cleaning. The level
of lubricant is already much better than on the outer layer of wires, although somewhat
flaky and denatured. The cleaned wires show very little residual galvanising, and a slight
roughening of the surface indicates the onset of corrosion.
As with the tug side sample, further 'into' the strand the nature of the lubricant and level
was better, and the wires had retained more of their galvanic protection (Figures 10.44 and
10.45). No corrosion was noted in the third and core layers.
As with the IWRC, the outer strands were completely dismantled to check for broken wires.
A total of nine outer wire breaks were found. There were no wire breaks in the second, third
or core layers.
before cleaning
after cleaning
Figure 10.43: Bridle sample main strand second layer of wires.
55
before cleaning
after cleaning
Figure 10.44: Bridle sample main strand third layer of wires.
before cleaning
56
after cleaning
Figure 10.45: Bridle sample main strand (some of) the third layer of wires and core wire (bottom).
10.4 Measurements made on strip down samples

In addition to the visual inspection and the mechanical tests on the wires (which will be
reported in the next sections, basic dimensions diameter and lay length of the wire rope
samples were checked (Table 10.1). It can be seen that the average diameters were quite
similar to the as manufactured value, as were the lay lengths. The bridle sample had the
greatest diameter reduction at 2.27%, but this is still within an acceptable value for this type
of construction [5, Table 5]. For completeness, Table 10.2 summarises the wire breaks
found.
Sample Lay length Average diameter % Loss average

[mm] [mm] diameter [mm]
As manufactured 534 78.46 -
Drum 540 78.55 - 0.11
Tug side 575 77.79 0.85
Bridle 540 76.68 2.27
Table 10.1: Summary of measured lay-lengths and wire rope diameters on inspected sections.
Outer strand
Sample IWRC core IWRC strands
outer wires
Drum 0 0 0
Tug side 59 62 4
Bridle 48 69 9
Table 10.2: Summary of broken wires found on inspected sections.
57
Figure 11.2: Tug side sample main failure, detailed view on one strand.
Figure 11.3: Composite SEM image of Wire 2 showing non tensile type failure.
59
Figure 11.4: SEM image of tensile (cup and cone) failure of Wire 6 (outer strand second layer large
wire).
Figure 11.5: SEM image of shear failure of Wire 4 (outer strand outer wire).
60
12 Mechanical tests on individual wires

Mechanical tests on the component wires of a wire rope are a useful means of assessing the
rope wire’s general condition. The results of such tests may be compared with the data from
the wire rope manufacturer (where available) and also with the relevant standard which
specifies minimum requirements.
It is immediately noted that the minimum standard requirements and manufacturers’ data
are for tests which are undertaken on the wire stock before it is spun into a wire rope. In the
work reported here, wires have been removed from the ex-service wire rope and cleaned
and straightened with care in order to allow testing. For the wire rope which is the subject of
this study, the main outer layer of the outer strands in the construction were subject to die-
forming (see Figure 2.1), which has led to non-circular wire cross sections. Previous testing
has shown that this can have quite a large effect on the results of the tensile strength tests,
and to a lesser extent an effect on the torsion test results, and the bending test results. It is
noted that the die-forming of the wire rope under discussion here is not as heavy as in
others, but it should be taken into account when interpreting the results of the tests. The
core wires (in the strands) and wires in the wire strand core should be relatively unaffected
by the die-forming and might be expected to give more consistent results.
Turning to the relevant standard, for the tests undertaken here, where possible, the results
will be compared with API 9A [1]. It is noted that the API standard does not specify
requirements for the reverse bend test, and so the results of these tests will be compared
with the criteria set out in BS EN 10264-2:2012 [6] which was effective at the time of the
wire rope manufacture.
The following types of mechanical tests were undertaken on wires from the wire rope
strands:
• Tensile tests;
• Torsion tests; and,
• Reverse bend tests
Attention was focussed on the outer strands and the outer wires of the outer strands, since
these represent the greatest percentage of load bearing area in the wire rope. Thus, for the
construction under consideration here it was decided to take:
• Three outer wires from each of four outer strands;
• One of each wire size from the second and third layers of three outer strands;
• One core wire from one outer strand;
• One wire of each size from the IWRC strand; and,
• One of wire of each size from the IWRC core.
Table 12.1 summarises the test samples which were extracted from the wire rope (one set of
each for the tension, torsion and bending tests).
61
Strand Wire position Number of tests
Outer strand 1* outer wire 3
2nd layer large and small 1+1
Third layer 1
Core 1
Third layer 1
Third layer 1
IWRC strand outer layer 1
core 1
IWRC core outer layer 1
core 1
TOTAL 26
*Note the strand numbers are completely arbitrary, strands were selected at random
Table 12.1: Summary of the samples taken from the wire rope for tensile, torsion and bend testing.
62
12.1 Tensile tests

Tensile tests were undertaken on individual wires removed from the strands disassembled
under §12. Especial care was taken when selecting samples from the very outer layers of the
wire rope to avoid those which might have been subject to undue mechanical damage (e.g.
abrasion).
Testing was conducted in accordance with ISO 10425:2003 [7] Annex B.
The wires were removed from the wire rope construction, labelled and cleaned. Following
cleaning they were straightened as much as practical by bending in the soft jaws of a vice or
through a specially designed jig. It is noted that whilst the wires were straightened as much
as possible, they were not perfectly straight. Hence it was not possible to measure their
elongation to failure as the initial part of the load-stroke relationship was affected by the
final wire straightening under load.
Once straightened, the wires were cut to the correct length for testing (nominal 450 mm).
The wires were tested in TTI Testing’s universal servo-hydraulic testing machine, with special
grips for wires, (Figure 12.1). The tests were conducted in stroke control with a test speed of
0.3 mm/s. (Appendix D presents the calibration certificate for this machine.)
Figure 12.1: Tensile testing of wire in 250 kN tensile testing machine.
63
12.2 Torsion tests

Torsion tests were undertaken on a set of wires as detailed in Table 12.1. The tests were
conducted in accordance with ISO 7800:2012 [8]. As is permitted by the standard, a gauge
length of 300 mm was used for all tests, the results being scaled to provide a number of
turns to failure for a length of 100dw (where dw is the wire diameter).
12.3 Reverse bend test

Reverse bend tests were undertaken on a set of wires as detailed in Table 12.1. The tests
were conducted in accordance with ISO 7801:1984 [9].
64
14 Results of the mechanical tests

Full results of the mechanical testing (tensile, torsion and reverse bend) are presented in a
series of Tables in Appendix E.
Considering first the results of the tests undertaken on the drum sample wires:
• The results of the tensile tests were very consistent, and very good. With the
exception of one wire (at 99%) all wires were in the range of 101% - 109% of the
average wire grade stated on the manufacturer's certificate (see Table 2.2 and
Appendix A). Wire breaks were typical cup and cone type failures, which was as
expected (Figure 14.1).
OS1 OW1 OS1 3L
OS1 CW IWRC CS CW
Figure 14.1: Typical appearance of wire failures in tensile tests on wires from the drum sample.
• The results of the torsion tests were a little more varied. The wires either passed by
some margin or failed rapidly (Figure 14.2). The torsion test is very sensitive to
surface condition, and it maybe that surface of the outer wires on some of the outer
strands was damaged leading to premature failure. The results of the tests on the
wires from the inside of the main strands were generally very consistent and good.
The tests undertaken on the IWRC wires were generally low. It is thought that the
reason for this may be the effort put into straightening the wires for testing. It is easy
to damage a wire despite taking great care in sample preparation. In contrast the
core wire which required no straightening performed well.
66
Figure 14.2: Appearance of wire failures in torsion tests on wires from the drum sample. (Top
OS OW1 (7 turns/100dw), bottom OS CW (27 turns/100dw)).
• The reverse bend test (in addition to the torsion test) is means of assessing a wire's
ductility. The API 9A [1] does not have a minimum requirement for wires tested in
this manner, so the requirements of Standard 10264-2 [6] have been used. All the
wires in the reverse bend tests satisfied the minimum requirements of the Standard.
This would suggest that the wire material was ductile, but in the torsion tests the
surface had been damaged.
Turning to the tug side sample:

• The results of the tensile tests on the outer strand outer wires were consistently all a
bit low (one at 83%, others at 92% to 99%), (Figure 14.3). The results of the tests on
the main strand inner wires were all very good (100% to 109%). The IWRC wires were
in the range 97% to 104% of the assumed original grade.
OS1 OW2 OS1 CW
Figure 14.3: Typical appearance of wire failures in tensile tests on wires from the tug side sample.
67
• The results of the torsion tests on the outer strand outer wires were all very low,
typically about 10% - 30% of the standard minimum requirement. The results of
these tests were significantly affected by the surface damage (wear) and corrosion
noted on the sample during the inspection (Figure 14.4). In contrast, all the wires on
the inside of the main strands performed well. These wires had not experienced
surface damage and were not corroded. The IWRC wires were very poor, but this is to
be expected given the state of the IWRC (two wire sizes could not be tested as a
sufficiently long sample could not be obtained!).
Figure 14.4: Appearance of wire failures in torsion tests on wires from the tug side sample. (Bottom
OS OW1 (3 turns/100dw), top OS CW (27 turns/100dw)).
• The reverse bend tests were a little better, with about half the outer strand outer
wires meeting the Standard minimum requirements. The outer strand inner wires
performed well and were at a similar level to the drum sample wires. The IWRC wires
performed better than in the torsion test, but the strand wires were still a little
down.
Finally the bridle sample:

• The initial inspection had shown that the condition of the bridle sample was similar
to that of the tug side sample. If anything the results were a little lower for the bridle
sample. The outer strand outer wires were affected by surface abrasion (Figure 14.5),
leading to results of 73% - 90% of the original strength. The wire rope wires from
inside the outer strands which were not affected by abrasion all performed well
(100% to 108%) as did the IWRC wires (100% to 104%).
68
OS2 OW3 - note surface abrasion OS3 OW2
OS3 2LL OS1 CW
Figure 14.5: Typical appearance of wire failures in tensile tests on wires from the bridle sample.
• The results of the torsion tests followed the familiar pattern - the outer strand outer
wires were low, typically 10% - 30% of the standard minimum requirement. The
outer strand inner layers of wires were comfortably in excess of the minimum, whilst
the IWRC wires (where available) performed poorly.
Figure 14.6: Appearance of wire failures in torsion tests on wires from the bridle sample. (Bottom
OS1 OW3 (3 turns/100dw), top OS1 CW (30 turns/100dw)).
69
• The reverse bend tests were a little better, with half the outer strand outer wires
meeting the Standard minimum requirements. The outer strand inner wires
performed well and as with the tug side sample, were at a similar level to the drum
sample wires. The IWRC wires performed better than in the torsion test.
70
15 Results of the zinc coat weight measurement

Full sets of results for the three samples are presented in Appendix F. For ease of reference
the results have been plotted in a series of bar charts (Figures 15.1 to 15.3).
With reference to Figure 15.1, in addition to the results of the tests, the chart also shows the
original level (from the Manufacturer's certificate (Appendix A and Table 2.2), as well as the
API 9A [1] Class B requirement.
It may be seen that as expected, the average coat weights on the drum sample are at the
Class B requirements.
The outer wires of the outer strands of both the tug side and bridle samples are both well
down. Remember that the coat weight measurement is an average figure, and does not
distinguish any distribution - so a wire may be corroding on an exposed section and
protected on the section towards the inside of the wire rope.
The relatively protected main strand inner layers had good levels of zinc, while the coatings
on the IWRC wires were very low. These findings are in line with those of the visual
inspection.
71
Manufacturer's original
Class B
Outer strand outer wires 2LL 2LS 3L CW IWRC OS IWRC CS

OW CW OW CW
Figure 15.1: Measured average residual galvanising on wires from various positions in the wire rope construction on the drum sample.
72
Class B

OW CW OW CW
Figure 15.2: Measured average residual galvanising on wires from various positions in the wire rope construction on the tug side sample.
73
Class B

OW CW OW CW
Figure 15.3: Measured average residual galvanising on wires from various positions in the wire rope construction on the bridle sample.
74
Figure 16.2: Showing for a range of loads how the catenary sag varies for a fixed line length (of
740 m) and weight per unit length (taken as 22.9 kg/m in water). Note only half the
catenary has been shown as it is assumed symmetrical about the y-axis.
Figure 16.3: Variation in horizontal load as a function of the span (wire rope length 740 m).
76
ΔF
Young’s modulus E is defined as: E = A
Δ

Where: ΔF is the change in (wire rope force);
A is the nominal wire rope area;
ℓ is the initial length of the sample; and,
Δℓ is the change in length caused by the change in load (ΔF)
Re-arranging this equation gives:
 ΔF 
EA =  ⋅
 Δ 
In this equation the term ΔF/Δℓ is the slope on the graph (shown by the blue line in Figure
16.5). The slope of the blue line in Figure 16.5 is 49.8 kN/mm and the length ℓ is 3190 mm,
which gives a value of EA of 159 MN.
16.2.2 Definition of the model

Δ ⋅ EA
Re-arranging the equation above gives: ΔF = .

• EA = 159 MN (characteristic wire rope axial stiffness)

• ℓ = the length of wire rope paid out,
• Assuming that ALP Forward and Transocean Winner are forced apart by wave
motions, and that the required stretch in the wire rope to accommodate the
movement apart Δℓ, is determined by the significant wave height Hs (i.e. Δℓ = Hs).
This permits calculation of the change in force ΔF caused by the sudden wave imposed
movement Δℓ.
Hence the maximum load in the wire rope is the change in load + the initial load
It is again noted that this calculation is approximate, but does give an idea of the levels of
load which might have been experienced in the wire rope during the storm.
79
Scenario 1 Scenario 2 Scenario 3
scope [m] 568 568 740
Hs = Δl [m] 8.4 16.8 5.8
initial load [tonnes] 120 120 220
EA [MN] 159 159 159
ΔF [tonnes] 240 479 127
Peak tension [tonnes] 360 599 347
Table 16.1: Approximate calculation of peak tensions experienced by the wire rope.
16.3 Fatigue loading

Another factor which should be considered is that the prolonged exposure of the wire rope
to significant varying loads during the storm will have inevitably caused some reduction in
endurance through fatigue. API RP 2SK [13] provides guidance for calculating the nominal
tension fatigue life of wire ropes of the size and construction under consideration here.
where:
N = Number of cycles
R = Ratio of tension range to rope MBL
M = 4.09 for six strand rope
K = 10(3.20 - 2.79Lm)
Lm = Ratio of mean load to rope MBL
Figure 16.7 shows the predicted number of cycles to failure for a rope loaded from zero to
peak tension (the load range) for varying load ranges. By way of example, it may be seen
that for a peak tension of 240 tonnes, the predicted number of cycles to failure would be
5,745 cycles. At a period of 14 s, failure would occur after about 22 hours.
81
Figure 16.7: Mean cycles to failure as a function of load range based on API RP 2SK [13].
82
It is considered that the outer strand outer wires have additionally been affected by the
onset of corrosion - the lubricant was generally missing or low/denatured. It is noted that
had the wire rope been assessed under the criteria of ISO 4309:2010 [5], then it would have
been subject to immediate discard.
In fact the condition of the wire rope is so poor as to raise the question of its condition when
last inspected in July 2016 [4]. It is noted that the wire rope was re-socketed in July 2016 -
given the condition of the bridle sample IWRC when terminated for break test, it might
prove informative to remove and examine the spelter socket on the wire rope.
Turning to consider the sorts of loads which the wire rope might have experienced during
the storm, an approximate calculation assuming a maximum wave height of 16.8 m indicates
a load of the order of 599 tonnes. This load is well in excess of the measured tug side sample
BL of 389 tonnes, and above the wire rope Fmin (485 tonnes) and the measured ABL of
494.1 tonnes. Given this estimate, it is not surprising the wire rope failed. It is quite possible
that the wire rope would have failed even if it had been as new.
It is noted that the load stated above assumes that the winch drum would not have slipped
or rendered. Details of the winch drum are not known, but a data sheet (Figure 17.2)
indicates that with 740 m paid out the brake holding load was about 1725 kN or 176 tonnes.
It has been reported that the winch had been rendering and loads of 180 - 220 tonnes had
been recorded [3]. It is possible that a shock load might be applied to the wire rope so
quickly that the drum did not have time to accelerate.
In addition to setting the winch to render, another option to help manage the loads on the
wire rope would be to extend the scope of the towing line. Whilst it is not ideal to drag a
wire rope along the seabed, in cases of extreme need it would be preferable to do this so as
to help avoid breaking the line.
In conclusion, this study has shown that the 77 mm wire rope installed on the ALP Forward
in May 2014 satisfied the requirements of API 9A [1]. Although the tow wire log [4] states
that the line is regularly washed and greased during recovery, it is far from clear that this
process is effective - the lubricant on the tug side and bridle samples were very low and the
wire rope has started to corrode. (Generally speaking it is preferable to avoid re-greasing a
wet wire rope as this can trap moisture in the wire rope accelerating corrosion. It is possible
that this has happened to the wire rope examined here.) The condition of the core was very
poor.
It is likely that the rope had become further degraded through fatigue damage during the
earlier part of the storm, and finally broke due to a one off overload.
Breaking load tests on sections of wire rope from either side of the main failure suggest that
at the time of the incident the wire rope strength was down by 21.3% to about 389 tonnes.
84
Figure 17.2: ALP Forward main winch data sheet (supplied by MAIB).
85
18 References
1 API 9A Specification for wire rope, 26th edition, May 2011, Reaffirmed April 2016,
published American Petroleum Institute.
2 Usha Martin Wire Rope Hand Book Published by Usha Martin (ushamartin.com) p75.
3 Pers. Comm. data supplied by MAIB (Capt. Emma Tiller) in .xls spread sheet 'Towline'.
4 ALP Maritime Services, Main tow-wire log .xls spread sheet forwarded by MAIB file
name 'Updated Tow Wire Record Alp Forward.xls'.
5 BS ISO 4309:2010 Cranes – wire ropes – care and maintenance, inspection and discard,
International Standards Organisation, 4th Edition, August 2010, ISBN: 978 0 580
59980 4.
6 BS EN 10264-2:2012 Steel wire and wire products – Steel wire for ropes Part 2: Cold
drawn non alloyed steel wire for ropes for general applications, British Standards
Institution, 2012.
7 ISO 10425:2003 Steel wire ropes for the petroleum and natural gas industries –
minimum requirements and terms of acceptance, International Standards
Organisation, August 2003.
8 ISO 7800:2012 Metallic materials – Wire – Simple torsion test, International Standards
Organisation, March 2012.
9 ISO 7801:1984 Metallic materials – Wire – Reverse bend test, International Standards
Organisation, May 1984.
10 BS EN 10244-1:2001 Steel wire and wire products – non-ferrous metallic coatings on
steel wire – Part 1: General principles, British Standards Institution, 2001.
11 Ractliffe, A.T. and Parsey, M.R. Man-made-fibre ropes for marine use, Trans North
East Coast Institution of Engineers and Shipbuilders (NECIES) 101 (1985) pp 183-
197.
12 Stretch, R. Met Office Marine Weather Data report compiled for MAIB Reference
msc/08/16/056, 30th August 2016, §7.2.
13 API RP 2SK Design and analysis of stationkeeping systems for floating structures, 3rd
edition October 2005, Addendum 2008, published American Petroleum Institute.
86
Appendices
Appendix A - Tow wire inspection certificate

Appendix B - Calibration certificate for the 14,000 kN test machine
Appendix C - Certificates for the break load tests
Appendix D - Calibration certificate for 250 kN tensile machine
Appendix E - Results of the mechanical testing on rope wires
Appendix F - Results of the average zinc coat weight measurement
87
Appendix A - Tow wire inspection certificate
88
89
Appendix B - Calibration certificate for the 14,000 kN test machine
90
91
92
93
94
95
Appendix C - Certificates for the break load tests
96
97
98
Appendix D – Calibration certificate for 250 kN tensile machine
99
100
101
Appendix E - Results of the mechanical testing on rope wires

This appendix presents the results of the mechanical tests on the wires taken from the drum,
tug side and bridle samples. Where appropriate, wires were taken from different strands
(numbered arbitrarily as described in Section 12 above). Table E1 below summarises the
designation used in the results Tables to identify wires from the different positions in the
wire rope construction.
Description of wire position in rope Wire designation in results Tables
Outer Strand (#) Outer Wire (#) OS# OW#
Outer Strand (#) 2nd Layer Large OS# 2LL
Outer Strand (#) 2nd Layer Small OS# 2LS
Outer Strand (#) 3rd Layer OS# 3L
Outer Strand (#) Core Wire OS# CW
IWRC Outer Strand, Outer Wire IWRC OS OW
IWRC Outer Strand, Core Wire IWRC OS CW
IWRC Core Strand, Outer Wire IWRC CS OW
IWRC Core Strand, Core Wire IWRC CS CW
Table E1: Description and short designation of the wire samples used in the results Tables.
In each case, the results of the tests on the wires have been compared to the original
measured value on the wire stock (where known) or in the absence of this information to
the minimum requirement specified by the appropriate standard. In order to aid a 'quick'
interpretation of the results, on the right hand side of each table is a column with coloured
cells. These cells are coloured as follows:
• green where the test result met or exceeded the original (or standard minimum)
value,
• orange if they were at or above 90% but below 100%; and,
• red if below 90%.
102

Inspection and Testing of Failed Tow Wire Rope PDF

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Inspection and Testing of Failed Tow Wire Rope PDF

Uploaded by

Copyright:

Available Formats

TTI Testing Ltd

Unit 2, Hithercroft Road, Wallingford, Oxfordshire, OX10 9DG, UK

Inspection and testing of failed tow wire rope

TTI Reference number TTI-IMLR-2016-5284

Date Rev. Description Prepared by Authorised by

This report presents the findings of a study undertaken on a wire rope

11 Inspection of tug side failure ..................................................................................... 58

ABL (measured) Actual Breaking Load

As a secondary issue, the emergency towing arrangement on Transocean Winner included a

Wire diameter, Tensile strength Bend (Average) Torsion Weight of zinc

[mm] [N/mm2] [-] [turns /100 dw] [g/m2]

4.39 2164 N/A 13 161

2.73 2096 N/A 23 142

3.53 2128 N/A 17 146

3.62 2075 N/A 16 146

4.81 2087 N/A 12 163

4.1 Delivery of the drum sample

Figure 4.1: Drum sample as received at TTI Testing.

4.2 Delivery of the tug side and rig side samples

Figure 4.5: Rig side sample in container (unwrapped).

Rig side sample

Tug side sample

5 Initial inspection of samples

5.1 Initial inspection of drum sample

Figure 5.1: General condition of the drum sample as received.

5.3 Initial inspection of rig side sample

6 Initial cutting schedule

6.1 Cutting schedule for drum sample

6.2 Cutting schedule for tug side sample

6.3 Cutting schedule for rig side sample

7 Delivery of bridle sample

8 Initial inspection of additional sample

8.1 Initial inspection of bridle sample

8.2 Proposed cutting schedule for bridle sample

• A 1 m length at 7 m - 8 m for a full strip down inspection and mechanical tests on

9 Wire rope breaking strength measurement

9.1 Test equipment

9.2 Results of the breaking load tests

Drum 4,493 94.4 92.7

Tug side 3,815 80.2 78.7

Bridle 3,707 77.9 76.5

Figure 9.3: Drum sample in test machine after testing.

Figure 9.4: Drum sample failure clear of termination.

Figure 9.5: Drum sample detailed view on failure region.

Figure 9.7: Tug side sample after testing, failure mid-length.

Figure 9.11: Bridle sample after testing, failure mid-length.

10 Strip down inspection of 1 m samples

10.1 Strip down inspection of Drum Sample

Figure 10.1: General condition of the drum sample.

Figure 10.5: Drum sample outer strand - second layer of wires.

Figure 10.6: Drum sample outer strand - third layer of wires.

Figure 10.9: Drum sample IWRC.

Figure 10.10: Drum sample IWRC outer strand cleaned.

Figure 10.12: Drum sample IWRC core strand.

Figure 10.13: Drum sample IWRC core strand cleaned.

10.2 Strip down inspection of the tug side sample

10.3 Strip down inspection of the bridle sample

Figure 10.30: General view on bridle sample.

Figure 10.39: Bridle sample IWRC outer strand (after cleaning).

10.4 Measurements made on strip down samples

Sample Lay length Average diameter % Loss average

Table 10.2: Summary of broken wires found on inspected sections.